Adjust an estimate according to assignee
- Jonathan Muse
- Cristina Szilágyi
- Amelie Winkler
Problem
You want to simplify estimation of difficult tasks when using Agile frameworks. For example, you are unsure which team member will pick up a task, and you can only assign it a rough estimate. In the event of new team member, estimation might be made even more difficult given the learning and adaptation curve. Thus, completion time may vary from one team member to another.
Solution
To solve this problem, one option is to customize the estimate.
For example, consider a task estimated at 1 day. While one team member may need 1 day 4 hours to resolve it, another might be able to resolve it in 6 hours.
By keeping track of the error margin for each user, and the average of issues worked on, you can ensure a better starting estimate for upcoming tasks. If a user is consistently taking longer than the average by, for example, 20%, then you can anticipate and raise the estimate by 20% for them.
To do this, add two post functions:
Start Progress
When a user starts progress on an issue, the remaining estimate is recalculated to correspond to their velocity. A user's velocity is the factor by which the original estimate is multiplied to ensure a more accurate approximation.Resolve Issue
When an issue is resolved, add the difference between the time spent working on the issue and the calculated amount at progress start. Then include this result in future approximations.
Keep track of the parameters by storing them as user properties. The properties take the following two values:
User velocity
Iterations number (Each time you alter the velocity, you increment this value).
Each user starts with a velocity of 1 and 0 iterations.
Step 1 - Progress start and estimate recalculation
When a user starts progress on an issue, multiply the original estimate by user velocity. Include the result in the remaining estimate.
This step is recommended only if no time was logged on the issue, as some users work on an issue intermittently.
It is also recommended (with minor adjustments) when the issue is re-assigned. To allow some flexibility, write all the necessary functions in a separate file that you can retrieve from other SIL™ scripts.
Start progress post function
include "common_velocity.incl"; // our "auto-estimate library"
updateRemaining(key); Step 2 - Resolving an issue and velocity recalculation
When an issue is resolved, calculate the approximation error and include it in the velocity. You can use it for future estimates.
Resolve issue post function
include "common_velocity.incl"; // our "auto-estimate library"
adjustVelocity(key); Step 3 - Add the Environment configuration
Use the SIL™ Environment to store the default velocity (1) as well as the user properties containing the velocity and iterations number.
sil.properties
default.velocity = 1
user.velocity = jjup.autoestimate.user.velocity
velocity.samples = jjup.autoestimate.velocity.samplesSIL Environment variables have been replaced with the more versatile persistent variables feature. Unlike environment variables, these variables can be set in the code. Also, they can be specific to each issue, based on issue context.
Step 4 - Add the library
To include the logic in one file and use it again when re-estimating (for example, in listeners or services), save the following file to your silprograms folder.